EP3640638A1 - Automatic optimization of liquid chromatography autosampler - Google Patents

Abstract

A device comprising an autosampler coupled to a liquid chromatography (LC) system, an electronic memory storage device (209), a data line, an autosampler control process and and independently operating LC control process, both implemented within a computer system or other controller system that is electronically coupled to both the autosampler module and to the LC module. The autosampler control process is adapted to read, from the electronic memory storage, a stored value of a pre-injection time (t _rj(k)) corresponding to the time (t _pj(k), t _qj(k)) required for the autosampler to perform operations (p _j(k), q _j(k)) preparatory to injecting a sample; calculate an autosampler delay time (Δt _rj ^sj(k)) from the stored pre-injection time value and from a sample analysis time(t _sj(k)); perform the autosampler operations preparatory to injecting the sample after delaying said autosampler operations for the autosampler delay time; if no autosampler or LC-apparatus errors or malfunctions have been encountered during the performing of said autosampler operations, cause the autosampler to inject the sample from the autosampler into the LC system after receipt of a signal from the LC system by the autosampler.

Description

Field of the Invention
• This invention pertains generally to liquid chromatography, and more particularly to methods for automatically operating an apparatus for delivering samples of liquid to be analyzed in a liquid chromatographic column.
Background of the invention
• In liquid chromatography, a compound is broken down into its components in a chromatographic column, so that these components can be further processed or analyzed by a detector, such as a mass spectrometer. In various liquid chromatography techniques, such as High-Pressure Liquid Chromatography (HPLC), the components of a sample to be separated or analyzed are dissolved in a mobile phase liquid, termed an eluent, and then conveyed by that liquid to a stationary phase within one or more chromatography columns. HPLC analyses are employed in a wide variety of applications, such as drug discovery and development, environmental testing, and diagnostics. In HPLC systems, the chromatographic columns are interconnected to other components by fluidic systems. Such fluidic systems generally include an injection valve, possibly one or more auxiliary valves, various solvent and wash fluid reservoirs, together with various interconnecting fluid tubing lines which are used. Such a fluidic system may be used to supply the liquid, with dissolved sample, to a chromatographic column, and to transfer chromatographically separated dissolved components from the column to, for example, a detector. Selected pressures ranging from substantially atmospheric pressure to pressures on the order of ten thousand pounds per square inch are common in HPLC systems to force the liquid into the column. A detector interfaced to a chromatography system may analyze various samples in serial fashion.
• In many situations, there is a need to analyze a large number of samples by HPLC in an efficient manner. To facilitate efficient sample selection and processing, automated analysis systems are available which usually include automatic sample injectors or auto-injectors. Such auto-injectors, often referred to as autosamplers, hold a plurality of samples to be analyzed and are able to feed these in series into a liquid chromatographic analysis system. Typical auto-injectors include a plurality of sample reservoirs, a syringe or syringe-like sample transport system, and automation and computer control systems. The auto-injector system may also include the fluidic system components - the valves, fluid reservoirs, and interconnecting fluid tubing lines - described in the previous paragraph. Auto-injectors commonly mimic cumbersome manual injection methods in which a metered aliquot of a sample is aspirated from a desired sample reservoir into a transfer tubing loop under the action of a syringe pump. The aspiration process is often controlled by pulling back on a plunger or piston to create a negative pressure resulting in aspiration of the sample. Upon reconfiguration of the valve of valves of the fluidic system, the sample may then be transferred to the column. The term "autosampler" is sometimes used to refer to just the aspiration and sample transport portion of such a system.
• FIGS. 1A-1B illustrate a known chromatography analysis system including an auto-injector, adapted from that described in United States Patent Application Publication US2009/0145205-A1 , in the name of inventor Hochgraeber. FIG. 1A illustrates the system with the high pressure valve 1 configured in a LOAD position and FIG. 1B illustrates the same system with the valve configured in an INJECT position. The valve has six fluid ports, shown as ports 11, 12, 13, 14, 15 and 16. A high-pressure pump 40 that can supply a constant flow rate of solvent (supplied from a not-illustrated solvent reservoir) under high pressure is connected to port 15. In the switching position of the valve as drawn in FIG. 1A, this flow reaches port 14 through groove 25 of the valve 1, and then reaches a chromatographic column 41. A sample aspiration device, such as sample needle 44, is configured so as to dip into and withdraw a portion of sample from a sample container 43. The sample needle 44 is fluidically connected to port 12 of the valve 1. Instead of being moved into sample container, sample needle 44 can be moved into a waste container 38 to dispose of excess liquid. The needle may be directed so as to access various sample containers 43 or waste containers 38 under the operation of a computer controlled mechanical robot device 39, such as a robotic arm and supporting structures.
• Typically, the robot device 39 is operable so as to controllably and automatically position the needle 44 over any of the sample containers 43, waste containers 38, or other containers or ports. The robot device is typically also operable so as to vertically dip the sample needle 44 into a sample container 43 so as withdraw a sample portion and to subsequently lift the sample needle out of the sample container. A syringe or syringe pump 42 for drawing sample liquid is connected to port 11 of valve 1. The two remaining ports 13, 16 are externally connected to one another via a sample loop 50. Sample fluid can thereby be drawn from sample container 43 into sample loop 50 with the aid of syringe or syringe pump 42. The switching position of the valve as drawn is referred to as the LOAD position, since the sample material is being loaded into the sample loop.
• In order to feed the sample material into the high-pressure liquid stream, the valve 1 is switched over to a second switching position, which is shown in FIG. 1B. In this configuration, sample loop 50 is looped into the liquid path between pump 40 and column 41. The sample liquid previously 18 drawn into sample loop 50 is thereby transported with the liquid stream coming from pump 40 into column 41, where the chromatographic separation takes place. A detector 45, commonly a mass spectrometer, may be connected downstream of the column. The switching position of the valve as drawn is referred to as the INJECT position, since the sample material is being injected into the high-pressure liquid. The entire process may be repeated for a subsequent sample. Although the sample containers 43 are illustrated, in FIGS. 1A-1B, as glass vials, they could alternatively comprise wells of microtiter plates or any other form of sample container. Sample reservoirs may be sealed with a plastic film or metal foil, or a septum.
• FIG. 2 illustrates another known chromatography analysis system including a different commonly-used auto-injector configuration. The system shown in FIG. 2 comprises a separate modular autosampler system 2 that comprises aspiration, transfer and dispense capabilities. The robot system 39 of the autosampler system 2 shown in FIG. 2 is operable so as to aspirate a portion of sample from any one of the sample containers 43 and then dispense the sample portion to a sample injection port 7 of a modular chromatography system 9. The illustration shown in FIG. 2 is schematic; typically, the injection port 7 may be configured so that the sample needle 44 may be inserted into the injection port to as to form a leak-tight seal therebetween during injection of the sample portion. As in the chromatography systems shown in FIG. 1, the system of FIG. 2 comprises a multi-port valve 1. However, the fluid connections of the system of FIG. 2 are plumbed differently from those of the system of FIG. 1 partly because the syringe pump 42 is a component of the autosampler module 2. The valve configuration shown in FIG. 2 is a LOAD configuration in which the sample introduced into injection port 7 is delivered to sample loop 50. With this configuration, any fluid previously in the sample loop 50 is flushed out to waste container 8. In an alternative ELUTE configuration (not illustrated in FIG. 2), the grooves 21, 23, 25 of the valve 1 are realigned, by rotation of a rotor of the valve, such that a buffer solution or other solvent is caused to pass through the sample loop 50 under high pressure conditions provided by pump 40, such that the previously loaded sample portion is flushed, together with the buffer or solvent, out of the loop 50 and into column 41.
• The auto-injector systems illustrated in FIGS. 1-2 are basic systems which provide limited fluid-routing flexibility and limited ability to change solvents during the course of a separation. More-complex systems are known - comprising one or more additional valves, fluid reservoirs or columns - which provide additional necessary or required functionality. For instance, United States Patent No. 7,588,725 , in the names of inventors Ozbal et al., teaches an autoinjection system having a sample injection valve, a column control valve and a wash control valve. The sample injection valve has a first position which applies a reduced pressure to a sample sipper tube for aspirating a fluidic sample into the sample sipper tube, and a second position which delivers the fluidic sample to a sample supply loop. The column control valve has a first position which delivers the fluidic sample from the sample supply loop to a sample chromatography column, and a second position which reverses direction of fluid flow through the sample chromatography column to deliver the fluidic sample to a sample analyzer. The wash control valve has a first position which supplies a wash buffer solution to the sample chromatography column in a forward fluid flow direction, and a second position which supplies elution solvent to flush the sample supply loop. Fluidic circuits taught by Ozbal et al. provide the capability of passing a fluid over the insoluble matrix of a chromatography column in a first direction such that an analyte in the fluid binds to the insoluble matrix, and back-eluting an elution fluid over the insoluble matrix in a second direction opposite the first direction to output a sample that includes the analyte. Fluidic circuits taught by Ozbal et al. provide the capabilities of delivering wash buffer solution from the wash control valve to the sample chromatography column; delivering elution solvent from the wash control valve to the sample supply loop and aspirating a second sample fluid while simultaneously outputting a first fluid in the sample loop to an analyzer.
• United States Patent 6,635,173 , in the name of inventor Brann discloses a multi-column chromatography system which is illustrated herein as FIG. 3 of the present document. As can be seen therein, the system 50 contains an autosampler 51 which includes a plurality of injection valves 54, a plurality of pumps 56, a plurality of columns 58, and a selector valve 60 and a detector 62. Columns 58 may comprise a wide variety of columns useful for chromatographic analysis which can be used to direct a fluid sample into the entrance orifice of a given detector. For example, columns 58 may comprise high performance HPLC columns, capillary electrophoresis columns, gas chromatography columns, flow injection transfer lines, etc. In addition, although not shown, the system may also preferably include a port valve, positioned before the columns, which in the case of a single column system (one or more pumps and one or more columns) operates to load sample in one direction and elute in the opposite direction, as previously described in this document. In the case of a two column system, the port valve provides a similar function, and also provides a loop for eluting solvent.
• Each combination of pump 56, injection valve 54, and column 58 (together with any associated tubing, additional valves, additional pumps or additional columns) shown in FIG. 2 may be regarded as a separate chromatographic system operating in parallel with three other such chromatographic systems. In that regard, it can be appreciated that each chromatographic system is controlled by the computer controller 63 to ensure that samples are introduced by the autosampler to avoid overlap at the detector end, and to ensure maximum use of the detector's time as a detector. In that regard, the system 50 may be considered as comprising four (4) independent chromatography systems, wherein each system contains one (1) or more pumps and one (1) or more columns. Each of such independent chromatography systems may comprise two (2) pumps and one (1) column, such that one pump is devoted to loading the column with sample, and one pump for elution.
• FIG. 4, reproduced from the aforementioned United States Patent 6,635,173 , illustrates curves 70, 72, 74, and 76 showing the procedural benefit of controlled staggered/sequenced sample injections as may be performed using the system 50 (FIG. 3). As can be seen, the detector analyzes each curve in sequence. In such regard the detector herein functions to detect and report curve 70, while those samples responsible for curves 72, 74, and 76, although in the process of being eluted in the column, have not yet exited from the column. Such programmed chromatography sequencing is provided herein by a computer control device 63 (FIG. 3) which, upon consideration of when the target sample is likely to exit the column, adjusts the introduction of samples from the autosampler into the columns to sequentially deliver eluant containing sample for sequenced detection. In other words, the computer controller 63 considers the samples in the autosampler, and the input of information concerning their anticipated data-collecting window at the detector, and selects those samples from the autosampler for introduction into the system to maximize detector use.
• As indicated in the above discussion, processing of any single sample by a liquid chromatography (LC) system generally begins with an autosampler. Unless otherwise indicated herein, the term "autosampler" is used from this point forward in this document to refer to a modular autosampler, such as the modular autosampler 2 shown in FIG. 2. Unlike operational procedures run by other components, such as pumps and detectors, it is often difficult to determine beforehand the duration of an autosampler procedure. Steps do not necessarily have precise times associated with them. Furthermore, times can vary significantly system to system depending on instrument configuration. In general, such an autosampler executes a programmed procedure of the following general form:
1. (a) Pre-sample steps (steps executed prior to drawing a sample). This sequence may include wash steps of the syringe and injector, introduction of one or more air gaps (to separate different samples), and movement into position over the position of a sample vial or other sample container. Issues encountered during this phase need not affect sample results. In other words, any malfunctions or other errors encountered during this pre-sample phase will generally not damage or otherwise modify a sample in such a way so as to give incorrect analytical results, since the sample remains in its container. Thus, any such malfunctions or other errors occurring during this phase, if detected by a control system, can possibly be compensated for by simply aborting an existing procedure and re-starting it from the beginning, by switching operations to a backup or concurrently running autosampler or LC system or channel or perhaps, by raising an alarm which will instruct an operator to take corrective action. Spans of time denoting the execution of pre-sample steps are identified by reference symbols including the letter "p" in the accompanying drawings and associated text - for example, a span of time associated with the execution of pre-sample steps of the j^th sample analysis procedure is identified (see FIGS. 5A-5B) with a generalized reference symbol of the form p_j. The k^th such span of time is identified (see FIGS. 7A-7B) by a generalized reference symbol of the form p_j(k).
2. (b) Sample Transport. Sample transport begins with physically drawing the sample. This step includes physically moving an aspirated sample and may, optionally, include certain sample preparation steps - such as mixing with a reagent, centrifuging, etc. Any issue encountered during this phase may directly affect the sample being transported. Possible problems which may occur during this phase may include leakage of a sample out of a sample needle (for low-viscosity samples), sample evaporation or time-dependent degradation of the sample, either by exposure of the sample to air, to an unfavorable temperature environment or to ambient light. Accordingly, it is desirable to adjust the timing of commencement of this phase so that the sample is held in the needle or other transport device for no longer than is necessary. Spans of time denoting the execution of sample transport steps are identified by reference symbols including the letter "q" in the accompanying drawings and associated text - for example, a span of time associated with the execution of sample transport steps of the j^th sample analysis procedure is identified (see FIGS. 5A-5B) with a generalized reference symbol of the form q_j. The k^th such span of time is identified (see FIGS. 7A-7B) by a generalized reference symbol of the form q_j(k).
3. (c) System Synchronization ("Sync"). The autosampler may wait for confirmation that an LC system is ready to receive an injection. The sync may operate through either software or hardware mechanisms. Because some samples may be reactive or prone to degradation or loss as noted above, it is desirable to minimize such waiting time. These "Ready" signals are identified by reference symbols 81a-81d in FIGS. 5A-5B and reference symbols 85a-85c in FIG. 7A.
4. (d) Injection. The system injects the sample into the LC system and starts other LC devices, effectively transferring control of the sample. Each injection operation requires a brief but finite time period - these are represented by time periods noted as t _inject in FIGS. 5A-5B. The completion of injection is indicated by reference symbols 83a-83d in FIGS. 5A-5B and reference symbols 87a-87b in FIG. 7A. At such times, injection-completed signals may be transmitted from the autosampler the LC channel system.
5. (e) Post-Injection Operations. The autosampler may execute several other operations on the back end of the program. Spans of time denoting the execution of post-injection steps are identified by reference symbols including the letter "z" in the accompanying drawings and associated text - for example, a span of time associated with the execution of post-injection steps associated with the j^th sample analysis procedure is identified (see FIGS. 5A-5B) with a generalized reference symbol of the form z_j. The k^th such span of time is identified (see FIGS. 7A-7B) by a generalized reference symbol of the form z_j(k).
• Full time spans including all of the austosampler procedural steps (a)-(e) above are identified by reference symbols including the letter "r" in the accompanying drawings and associated text - for example, time spans r1, r2, r3 and r4 in FIGS. 5A-5B. FIG. 5A illustrates a first example of a conventional timing scheme for coordinating the operation of an autosampler and an LC system. The horizontal axis in FIG. 5A represents time. In FIG. 5A, as well as in FIG. 5B and FIG. 7, boxes at the base of the diagram represent periods of time during which a liquid chromatography channel - such as a liquid chromatography / mass spectrometry (LC/MS) system or a single channel thereof - is busy performing the operations of separating components of and possibly identifying or quantifying chemical species within a previously injected sample. Different patterns applied to boxes at the base of the diagram represent either different samples or different sample analysis procedures, denoted as s_i (where i is an integer) such as samples or procedures s1, s2 and s3. Boxes drawn in solid lines at the top of the diagrams of FIG. 5 and FIG. 7 - specifically, boxes r1, r2, r3 and r4 - represent periods of time that a robotic autosampler devotes to performing steps (a)-(b) as listed above; un-patterned boxes drawn in dotted lines represents auto-sampler idle time, either as the result of an intentional delay period or else during which the autosampler is waiting for the LC channel or system to become available.
• To improve timing between subsequent samples, a typical strategy, as illustrated in FIG. 5A, has been to start the autosampler as early as possible on subsequent samples. Thus, as may be seen in FIG. 5A, autosampler operational procedure r2, relating to preparation for injection of a sample whose analysis is noted at s2, commences immediately after injection of a prior sample whose analysis is noted at s1. Since the robot operations may require a time period of 1-3 minutes whereas the sample analyses may require substantially longer periods - for instance 4-12 minutes, this so called "Look-Ahead" methodology, may frequently lead to situations in which an autosampler has completed steps prior to injection and thus spends a significant period of time waiting for the "System Sync" to continue. Such waiting periods are denoted by the time intervals Δt _r1, Δt _r2 and Δt _r3 in FIG. 5A. The existence of such waiting periods can have several negative consequences: (i) samples sensitive to temperature or other factors may be affected by the length of time spent in the syringe and also by the potential time variations seen between samples; (ii) samples with very low viscosity may begin to drip or mix across air gaps; and (iii) the probability of losing a sample due to an abort or error of a previous sample is increased.
• A second conventional timing scheme for coordinating the operation of an autosampler and an LC system is illustrated in FIG. 5B. The procedure illustrated in FIG. 5B attempts to overcome the above-noted fact that it is preferable to minimize the time a sample spends in the dispensing syringe or needle by attempting to start the auto-sampler such that it is ready to make an injection - that is, the "pre-injection" steps (a) and (b) are completed - just prior to receipt of a "System Sync" signal. This method utilizes a timing-data database 206 that contains autosampler timing estimates entered manually by users based on record-keeping of the time required to conduct autosampler operations during previous analytical runs. Since the various autosampler operation times may vary between different samples or different sample analysis procedures, such as samples or procedures s1, s2 and s3, different timing estimates should be maintained, in parallel, for each of the various samples or procedures. Also the times required to undertake the various LC sample analyses are expected to vary between the different samples or procedures s1, s2 and s3, etc., as is indicated by the varying widths of the differently patterned boxes along the time axis.
• The time estimates in the database 206 (FIG. 5B) are used to calculate predetermined autosampler "lag times" that measure the time between the injection of a sample into the LC system and prior to the commencement of the sequence of autosampler steps associated with the next sample. Thus, for instance, in FIG. 5B, the quantity ${\mathrm{\Delta }\mathit{t}}_{r2}^{\mathrm{<figure-callout\; id="1"\; label="s"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">s</figure-callout>}1}$

  

  represents the predetermined autosampler lag time that measures the time interval from the commencement of execution of the LC sample or procedure s1 until the beginning of the sequence of autosampler steps, r2, required to execute pre-sample steps (a) and sample transport steps (b) in preparation for injecting the sample associated with the next analysis indicated at s2. Each "lag time", so defined, is related to, but not necessarily identical to, a "delay time" which is a wait time inserted between the completion of autosampler post-injection steps and the commencement of the sequence of autosampler steps associated with the next sample. Such delay times, so defined, are indicated by the time segments outlined in dotted lines in FIG. 5B. As indicated in FIG. 5B, the correct calculation of the predetermined delay time or, equivalently, lag time, will cause the autosampler to finish its pre-injection steps (indicated by the shaded boxes p2 and q2) just about at the same time that the LC system transmits the System Sync signal 81b indicating that it is ready to receive the sample associated with the analysis indicated at s2.
• Similarly, the quantity ${\mathrm{\Delta }\mathit{t}}_{r3}^{s2}$

  

  (FIG. 5B) represents another predetermined autosampler lag time. This lag time measures the time from the commencement of execution of the LC sample or procedure s2 until the commencement of the sequence of autosampler steps, r3, required to prepare to inject the sample associated with the next analysis indicated at s3. In this hypothetical case, FIG. 5B indicates that the lag time and associated delay time (dotted line time segment after post-injection steps z2), as calculated from the user-estimated data, was too short, causing the existence of an unplanned autosampler idle time while the autosampler module waits for receipt of the System Sync signal 81c. The next predetermined autosampler lag time, ${\mathrm{\Delta }\mathit{t}}_{r4}^{\mathrm{<figure-callout\; id="3"\; label="s"\; filenames="imgf0003.png,imgf0005.png"\; state="\{\{state\}\}">s</figure-callout>}3},$

  

  (which in this instance is equivalent to the delay time) is inserted immediately after the commencement of execution of the LC sample or procedure s3 and prior to beginning the sequence of autosampler steps, r4, required to prepare to inject the sample associated with the next analysis indicated at s4. In this particular hypothetical case, FIG. 5B indicates that the lag time, as calculated from the user-estimated data, was too long, thereby causing LC system to idly wait for a period time while the autosampler completes its pre-injection steps after the LC system signaled, at 81d, that it was ready to receive the sample. Such delays in the operation of the LC system can lead to overall analytical inefficiency which can become significant during the running of numerous samples of a batch.
• The procedure illustrated in FIG. 5B requires diligence and conscientiousness by users to enter timing settings and to adjust them whenever necessary. Such timing setting adjustments should generally be entered whenever an LC analytical procedure or an autosampler procedure is modified. Typically, these procedures may both be modified in concert. Timing adjustments should also be entered whenever an analytical procedure is transferred to another system, since the timings are generally machine-specific. Failure to make such adjustments can seriously affect multiplexing timing efficiency and, as noted above, can effect overall system efficiency or can cause degradation or loss of samples. Moreover, robotic sample aspiration, transport and dispensing operations and multiplexed chromatography systems (FIGS. 1-3) are typically associated with systems designed for automated screening operations, such as high-throughput screening (HTS) systems involving virtually continuous sequences of analyses and large numbers of samples. Such automated high-throughput systems are often utilized, for instance, in drug-discovery procedures and may be used for clinical applications such as drug-testing or screening for vitamin deficiencies or for diagnostic biomarkers. Such automated screening systems are frequently designed for unattended or overnight operation with little operator intervention other than initial loading of samples. In such systems, sample containers may be accessed in a "random" fashion which is generally unpredictable to a user - as may be the case of the sample access and analysis operations are under the control of a computerized scheduler, as may occur with multiplexed chromatographic systems. Because of these factors, it may be difficult or impossible, in practice, for a user to manually maintain detailed logs of elapsed autosampler times for different sample types or procedures.
Summary of the Invention
• In accordance with the present teachings, software automatically tracks various timing characteristics of each analytical procedure in use in an LC system coupled to a modular autosampler. The software uses these timing characteristics in a predictive model to start an autosampler at the optimum time; such that little to no time is spent waiting for system synchronizations. As samples are run, software continually tracks "Pre-Sample" and "Pre-Injection" times for each sample. The software maintains a record of the longest times encountered for each parameter of each method in use. Time spent waiting for synchronization signals is not considered when generating these timing records. Also, if a deviation is encountered (such as a hold, error, or abort event) during analysis of a particular sample, all or part of the method timing parameters relating to that sample are discarded. Using timing data obtained from recently performed chromatography experiments, the most current values of longest time delays are used to predict when the autosampler should begin executing its programmed sequence of operations. Generally, this will occur when the time remaining on a previous sample run is equivalent to the total Pre-Injection time for a pending auto-sampler procedure. With multiplexing, the predictive methods taught herein can additionally use the Pre-Sample time to best optimize "Detector Reset" delay times.
• The present invention provides several important benefits. In a single channel system, variations in the time that a particular sample may be held are minimized and, consequently, the risk associated with holding the temperature of or otherwise excessively handling a sensitive sample is minimized. Additionally, the risk of losing or damaging a sample being transported by the autosampler due to an error or abort elsewhere in an autosampler system is minimized. In a multiplexed system, method embodiments according to the present invention are able to automatically optimize multiplexing efficiency while maintaining maximum flexibility and robustness.
• Further aspects of the present disclosure as set forth in the following numbered clauses:-
• Clause 1. A method for optimizing the operation of a system comprising an autosampler and a liquid chromatography (LC) apparatus, wherein the LC apparatus is configurable to chromatographically separate components of a sample in accordance with one or more LC analysis procedures, wherein the autosampler is configurable to withdraw a sample portion from a container, prepare the sample portion for injection into the LC apparatus and transport the sample portion to the LC apparatus in accordance with the one or more LC analysis procedures, the method characterized by:
  • reading, from an electronic memory storage device, a stored pre-injection time value corresponding to the time required for the autosampler to perform operations preparatory to injecting the sample portion into the LC apparatus;
  • calculating an autosampler delay time from the stored pre-injection time value, a sample analysis time value and a time of an immediately preceding injection;
  • performing the autosampler operations preparatory to injecting the sample portion into the LC apparatus after delaying said autosampler operations for a time corresponding to the autosampler delay time; and
  • if no autosampler or LC-apparatus errors or malfunctions have been encountered during the performing of said autosampler operations, injecting the sample portion from the autosampler into the LC apparatus after transmission of a signal from the LC apparatus to the autosampler.
• Clause 2. A method as recited in Clause 1, further characterized in that the step of reading, from an electronic memory storage device, a stored pre-injection time value comprises reading a pre-injection time value that is the sum of a first stored time value and a second stored time value, the first stored time value corresponding to autosampler operations associated with preparing to withdraw the sample portion from a container and the second stored time value corresponding to autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion.
• Clause 3. A method as recited in Clause 1, further characterized by:
  • measuring an elapsed time value corresponding to the step of performing the autosampler operations preparatory to injecting the sample portion into the LC apparatus; and
  • replacing the stored pre-injection time value stored on the electronic memory storage device with the measured elapsed time value if the measured elapsed time value is greater than the pre-injection time value stored on the electronic memory storage device.
• Clause 4. A method as recited in Clause 2, further characterized by:
  • measuring a first elapsed time value corresponding to the autosampler operations associated with preparing to withdraw the sample portion from the container;
  • measuring a second elapsed time value corresponding to the autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion;
  • replacing the first stored time value stored on the electronic memory storage device with the measured first elapsed time value if the measured first elapsed time value is greater than the first stored time value; and
  • replacing the second stored time value stored on the electronic memory storage device with the measured second elapsed time value if the measured second elapsed time value is greater than the second stored time value.
• Clause 5. A method as recited in either of Clauses 2 or 4, further characterized in that if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations associated with preparing to withdraw the sample portion from the container, then execution of the currently executing operation and any subsequent autosampler operations associated with the sample portion are aborted.
• Clause 6. A method as recited in either of Clauses 2 or 4, further characterized in that if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion, then the sample portion is discarded.
• Clause 7. A method as recited in any of Clauses 1-4, further characterized in that:
  • the LC apparatus comprises a plurality of channels; and
  • if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations preparatory to injecting a sample portion into the LC apparatus, then the autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion are delayed until one of the plurality of channels is available to receive the sample portion.
• Clause 8. A method as recited in Clause 1, further characterized by:
  • reading the sample analysis time value from a stored value of the sample analysis time on the electronic memory storage device prior to the calculating of the autosampler delay time;
  • measuring an elapsed sample analysis time value corresponding to the operation of the LC apparatus during analysis of the sample portion; and
  • replacing the stored value of the sample analysis time stored on the electronic memory storage device with the measured elapsed sample analysis time value if the measured elapsed sample analysis time value is greater than the stored value of the sample analysis time.
• Clause 9. A method as recited in Clause 1, wherein the stored pre-injection time value comprises one of a plurality of stored pre-injection time values, each of the plurality of pre-injection time values corresponding to a different respective one of a plurality of LC analysis procedures.
• Clause 10. A method as recited in Clause 2, wherein the first stored time value comprises one of a first plurality of stored time values, each of the first plurality of stored time values corresponding to autosampler operations associated with preparing to withdraw a sample portion from a container in accordance with a different respective one of a plurality of LC analysis procedures.
• Clause 11. A method as recited in Clause 10, wherein the second stored time value comprises one of a second plurality of stored time values, each of the second plurality of stored time values corresponding to autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion.
• Clause 12. A system comprising an autosampler, a liquid chromatography (LC) apparatus, an electronic memory storage device electronically coupled to both the autosampler and the LC apparatus, a data line electrically coupled between the autosampler and the LC apparatus, an autosampler control process and an independently operating LC control process, wherein the LC control process is operable to cause the LC apparatus to chromatographically separate components of a sample in accordance with one or more LC analysis procedures and to transmit a signal to the autosampler control process upon completion of each LC analysis procedure, wherein the autosampler control process is operable to cause the autosampler to withdraw a sample portion from a container, prepare the sample portion for injection into the LC apparatus and transport the sample portion to the LC apparatus in accordance with the one or more LC analysis procedures, the system characterized in that the autosampler control process is operable to:
  • read, from the electronic memory storage device, a stored pre-injection time value corresponding to the time required for the autosampler to perform operations preparatory to injecting the sample portion into the LC apparatus;
  • calculate an autosampler delay time from the stored pre-injection time value, a sample analysis time value and the time of an immediately preceding injection;
  • cause the autosampler to delay operations for a time corresponding to the autosampler delay time;
  • cause the autosampler to perform operations preparatory to injecting the sample portion into the LC apparatus; and
  • if no autosampler or LC-apparatus errors or malfunctions have been encountered during the performing of said autosampler operations, injecting the sample portion from the autosampler into the LC apparatus after receipt of the signal from the LC apparatus.
• Clause 13. A system as recited in Clause 12, wherein the reading, from the electronic memory storage device, of the stored pre-injection time value comprises reading and summing together a first stored time value and a second stored time value, the first stored time value corresponding to autosampler operations associated with preparing to withdraw the sample portion from a container and the second stored time value corresponding to autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion.
• Clause 14. A system as recited in Clause 12, further characterized in that the autosampler control process is further operable to:
  • measure an elapsed time value corresponding to the performing of the autosampler operations preparatory to injecting the sample portion into the LC apparatus; and
  • replace the stored pre-injection time value stored on the electronic memory storage device with the measured elapsed time value if the measured elapsed time value is greater than the pre-injection time value stored on the electronic memory storage device.
• Clause 15. A system as recited in Clause 13, further characterized in that the autosampler control process is further operable to:
  • measure a first elapsed time value corresponding to the autosampler operations associated with preparing to withdraw the sample portion from the container;
  • measure a second elapsed time value corresponding to the autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion;
  • replace the first stored time value stored on the electronic memory storage device with the measured first elapsed time value if the measured first elapsed time value is greater than the first stored time value; and
  • replace the second stored time value stored on the electronic memory storage device with the measured second elapsed time value if the measured second elapsed time value is greater than the second stored time value.
• Clause 16. A system as recited in any of Clauses 12-15, further characterized in that:
  • the LC apparatus comprises a plurality of channels; and
  • the LC control process module is operable to:
    • determine if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations preparatory to injecting a sample portion into the LC apparatus; and
    • cause the autosampler control process to delay autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion until one of the plurality of channels is available to receive the sample portion, if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations preparatory to injecting a sample portion into the LC apparatus.
• Clause 17. A system as recited in Clause 12, wherein the stored pre-injection time value comprises one of a plurality of stored pre-injection time values, each of the plurality of pre-injection time values corresponding to a different respective one of a plurality of LC analysis procedures.
• Clause 18. A system as recited in Clause 13, wherein the first stored time value comprises one of a first plurality of stored time values, each of the first plurality of stored time values corresponding to autosampler operations associated with preparing to withdraw a sample portion from a container in accordance with a different respective one of a plurality of LC analysis procedures.
• Clause 19. A system as recited in Clause 18, wherein the second stored time value comprises one of a second plurality of stored time values, each of the second plurality of stored time values corresponding to autosampler operations associated with transporting of the sample portion or with preparing and transporting of the sample portion.
Brief Description of the Drawings
• The above noted and various other aspects of the present invention will become apparent from the following description which is given by way of example only and with reference to the accompanying drawings, not drawn to scale, in which:
• FIG. 1A is a simplified schematic representation of an autosampler according to prior art with a high-pressure injection valve and with connected fluidic components, wherein the valve is in the LOAD position;
• FIG. 1B is a representation analogous to FIG. 1A, but with the high-pressure valve in the INJECT position;
• FIG. 2 is a simplified schematic representation of a modular robotic autosampler system coupled to a modular liquid chromatography (LC) system;
• FIG. 3 illustrates in schematic view a known multi-column chromatography system;
• FIG. 4 illustrates the typical results of the multi-column chromatography system of FIG. 2;
• FIG. 5A is a schematic illustration of a first conventional method for coordinating the operation of an autosampler and an LC system;
• FIG. 5B is a schematic illustration of a second conventional method for coordinating the operation of an autosampler and an LC system;
• FIG. 6 is a schematic illustration of a preferred control software architecture for implementing methods in accordance with the present teachings;
• FIG. 7A is a schematic illustration of a method for optimizing the operation of an autosampler and an LC system in accordance with the present teachings;
• FIG. 7B is a schematic illustration of a variation to a method for optimizing the operation of an autosampler and an LC system in accordance with the present teachings;
• FIG. 8A is a flow diagram for a method for optimizing the operation of an autosampler and an LC system in accordance with the present teachings;
• FIG. 8B is a flow diagram of a portion of an alternative method for optimizing the operation of an autosampler and an LC system in accordance with the present teachings; and
• FIG. 8C is a flow diagram of a method for optimizing the operation of an autosampler and a multiplexed LC system in accordance with the present teachings.
Detailed Description of the Invention
• The following description is presented to enable any person skilled in the art to make and use the invention, and is provided in the context of a particular application and its requirements. Various modifications to the described embodiments will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. Thus, the present invention is not intended to be limited to the embodiments and examples shown but is to be accorded the widest possible scope in accordance with the claimed features and principles. To appreciate the features of the present invention in greater detail, please refer to FIGS. 1-8 in conjunction with the following discussion.
• FIG. 6 is a schematic illustration of a preferred control software architecture for implementing methods in accordance with the present teachings. The modular software architecture system 203 illustrated in FIG. 6 is useful for - but not limited to - control of the operation of a system comprising the modular hardware architecture shown in FIG. 2, with a modular autosampler fluidically and electronically coupled to a modular LC system, such as an LC/MS system. In this discussion, it is to be kept in mind that the hardware components illustrated in FIG. 2 are illustrated in schematic and simplified form and that any real system may comprise various additional components. The novel methods disclosed herein are intended to be applied generally to any system comprising coupled autosampler and LC modules. The modular software architecture illustrated in FIG. 6 is implemented within a computer system or other controller system that is electronically coupled to both the autosampler module and to the LC module (but which is not specifically illustrated in FIG. 2).
• The modular software architecture system 203 (FIG. 6) comprises an autosampler control process and an LC control process. Each of the autosampler control process and the LC control process may be implemented in its own separate module - i.e., an autosampler control process module 204 and a separate LC control process module 208 - each module containing code specific for executing its respective process. In the following discussion, references to each of these process modules can be replaced with references to just the respective process, without any loss of generality. The autosampler control process module 204 and LC control process module 208 function concurrently and mostly independently of one another but may pass timing synchronization signals and other information between them. Updated instrument timing data is maintained within a computer-readable memory module 209, such as a hard disk drive, flash memory or any other form of data storage. Either the autosampler control process module 204 or the LC control process module 208 may store timing data, other status data or other information on the readable memory module 209 so as to be subsequently read by the other of the modules. Low-level timing or synchronization signals may be passed between the modules by means of a data line 210.
• The autosampler control process module 204 may comprise instructions to mechanically control the horizontal and vertical location of an aspirating and dispensing needle or syringe, to control the position of a moveable tray or carousel that holds various sample containers temporarily housed in or on the autosampler, to control one or more valves which direct various fluids - as, for instance, washing solutions - through the syringe or needle, to control one or more pumps that are used to draw fluid into or dispense fluid out of the needle or syringe, to control the physical environment - for instance, temperature - of various fluid or sample reservoirs on or in the autosampler, among various other functions. The autosampler control process module 204 also comprises a timer capability to keep track of the time required for various autosampler operations and may also comprise instructions to read various sensors of the autosampler, such as position sensors, temperature sensors, fluid level sensors, etc.
• The LC control process module 208 may comprise instructions to, to control one or more valves which direct various fluids - as, for instance, buffer solutions, washing solutions, various mobile phases - through one or more columns of an LC instrument or to waste, to control one or more pumps that urge fluids into and through the various columns, sample loops and other tubing interconnections of the LC system, to control the physical environment - for instance, temperature - of various fluid reservoirs, sample reservoirs or columns of the LC system, among various other functions. The LC control process module 208 may further comprise instructions to control various operating parameters of a detector that receives separated chemical species from the LC system and to read information relating to the species from the detector.
• FIG. 7A is a schematic illustration of a method for coordinating the operation of an autosampler and an LC system in accordance with the present teachings, using a similar format to that provided in FIGS. 5A-5B, with time, t, progressing from left to right across the diagram. The illustration in FIG. 7A only considers the controlling of and the updating of records of timing data relating to one particular chromatographic sample analysis procedure, denoted as s1. For instance, the procedures s1 could comprise a particular assay, a particular assay applied to a particular class of samples or to a particular sample, etc. For purposes of illustration, the various instances of execution of the procedure s1 are indicated as occurring consecutively. Although such repetitive execution of a single analytical procedure may occur in many instances, different analytical procedures may be interspersed with one another in various other situations, perhaps in order of priority, in some other pre-determined order or, possibly, in random fashion.
• Because the conditions under which chromatography experiments are run are performed under tight constraints designed to maintain the experiments of constant duration, the sample analysis time, which is the time required to perform the particular LC analysis procedure is assumed, for purposes of this initial discussion, to be maintained at the constant value of t _s1. This quantity is the time between which sample injection is complete and the subsequent indication, by the LC system, that it is ready to receive another sample. If the quantity t _s1 cannot be assumed to be constant, then the variation of this quantity may be noted by the LC control process module 208 and recorded in the timing database (in addition to the autosampler times) so as to be incorporated into the calculations of autosampler delay times. Further, because each LC analysis procedure (e.g., procedures s1, s2, s3, etc., see FIG. 5) may be associated with its own unique, different time (e.g., t _s1, t _s2, t _s3, etc.) required to perform the respective procedure, the procedure for calculation of autosampler delay times may take into account not only the time required to perform the autosampler operations associated with preparing to inject the next sample but, also, the time required to perform the procedure that that is executed just prior to injection of the next sample into a liquid chromatograph. Accordingly, the k ^th instance of the lag time implemented prior to pre-injection autosampler functions r_i (preparatory for analysis procedure s_i) at a time that procedure sj is executing is herein denoted by the symbolism ${\mathrm{\Delta }\mathit{t}}_{\mathit{ri}}^{\mathit{sj}}\left(k\right)\mathrm{.}$

  
• Returning to the discussion of FIG. 7A, a first injection in preparation for a first execution (index k=1) of the s1 analytical procedure begins after a System Sync signal 85a indicating that the LC system is ready to receive a sample, after finishing a prior analytical procedure, s0. The various instances of execution of the autosampler pre-injection operations prior to running the s1 procedure are indicated by segmented boxes drawn in solid lines and are indicated, in sequence, by the symbols r1(0), r1(1) and r1(2). For the initial injection, it may happen that no prior timing data will be available in the timing database. Accordingly, in this example, ${\mathrm{\Delta }\mathit{t}}_{\mathit{r}1}^{\mathit{s}0}\left(\mathrm{i}\right)=0$

  

  -that is, no time delay is implemented prior to the first execution of the autosampler operations r1(0), which occur while the prior procedure s0 is being executed. Other situations in which the delay time may be set to zero may occur if the database has been corrupted, if the procedure has been modified or if the procedure has been ported to a new instrument.
• In the hypothetical example illustrated in FIG. 7A, the time duration of autosampler pre-injection operations (comprising pre-sample operations p1(0) and sample transport operations q1(0) of the autosampler) during the procedure r1(0) is measured to be t _r1(0). In general, the autosampler control process module maintains a record of the maximum previously encountered value of the time required for the autosampler pre-injection operations, denoted, in this instance as ${\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}^{\mathrm{}1},$

  

  which is given as ${\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}^{\mathrm{}1}=\mathrm{<figure-callout\; id="1"\; label="max\; t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">max</figure-callout>}\left\{t_{\mathrm{r}}\right\}\left\{{\mathrm{}1}_{}\left(0\right),{\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}\left(1\right),{\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}\left(2\right),\dots ,t_{\mathrm{r}1}\left(n\right)\right\}\mathrm{.}$

  

  In a practical sense, the autosampler control process module compares the most recently encountered value with the maximum value already in the database and, if the newly-measured value is greater than the prior maximum, simply replaces the old maximum value with the newly measured value. After the first execution of the autosampler pre-injection operations associated with the analytical procedure s1, the first time value, t _r1(0), is the maximum value and, accordingly, this value is recorded.
• When the LC system module provides the "Ready" signal 85a, the sample to be analyzed by the analytical procedure s1 is injected at 87a and the procedure s1 is initiated. The next instance of execution of autosampler pre-injection operations r₁(1) does not begin immediately, but instead occurs after a time delay corresponding to or determined from the lag time ${\mathrm{\Delta }\mathit{t}}_{\mathit{r}1}^{\mathit{<figure-callout\; id="1"\; label="s"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">s</figure-callout>}1}\left(1\right)$

  

  which is calculated, generally as ${\mathrm{\Delta }\mathit{t}}_{\mathit{r}1}^{\mathit{<figure-callout\; id="1"\; label="s"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">s</figure-callout>}1}\left(1\right)=t_{s1}-{\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}^{\mathrm{}1},$

  

  where the value of ${\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}^{\mathrm{}1}$

  

  is retrieved from the timing database. Using the maximum of all previously encountered pre-injection times ensures that the autosampler module will almost always commence operations early enough so as to complete its pre-injection tasks prior to the time at which the LC module is ready to receive a sample, thereby ensuring that the system operates efficiently. In this particular instance, the inserted time delay (the delay time) is simply the lag time ${\mathrm{\Delta }\mathit{t}}_{\mathit{r}1}^{\mathit{<figure-callout\; id="1"\; label="s"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">s</figure-callout>}1}\left(1\right)=t_{s1}-{\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_r\left(0\right)\mathrm{.}$

  
• In this hypothetical example, the time duration required to execute the autosampler pre-injection operations is not constant and, the time duration, t _r1(1), of autosampler pre-injection operations of the procedure r₁(1) is greater than the time duration, t _r1(0), of the initial execution of this operational sequence. As a result, the execution of autosampler pre-injection operations p₁(1) and q₁(1) is seen to extend beyond the time - at System Sync signal 85b - at which the LC system is ready to receive a sample, thereby causing the LC system to wait until the injection is subsequently completed, later than expected, at 87b. Because, in this case, ${\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_r\left(1\right)>{\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}^{\mathrm{}1}={\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_r\left(0\right),$

  

  the newly encountered time duration t _r1(1) becomes the new maximum value recorded in the database. The subsequent autosampler lag time, ${\mathrm{\Delta }\mathit{t}}_{\mathit{r}1}^{\mathit{<figure-callout\; id="1"\; label="s"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">s</figure-callout>}1}\left(2\right),$

  

  is calculated using this new value of ${\mathrm{<figure-callout\; id="1"\; label="t\; r"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{r}}^{\mathrm{}1}\mathrm{.}$

  

  In this hypothetical example, this subsequent lag time and its corresponding time delay are correctly calculated such that the next autosampler pre-injection operation sequence comprising operations p₁(2) and q₁(2), terminates at the expected time, when the LC system provides the next System Sync signal 85c.
• As discussed previously, the pre-injection steps referred to in the above discussions comprise initial pre-sample steps (a) followed by subsequent sample transport steps (b). Accordingly, the pre-injection time may be decomposed into: (a) a pre-sample time during which all preparatory steps prior to withdrawing a sample from a sample container are executed and (b) a sample transport time during which the aspirated sample in a needle or syringe is moved from to an injection port. FIG. 7B illustrates in detail the events in the vicinity of the k^th execution of the LC analytical procedure sj(k) and the concurrent autosampler procedure r_j(k) having a pre-injection time duration of t _rj(k). As may be seen in FIG. 7B, this time duration is comprised of a pre-sample time component, t _pj(k) and a sample transport time component t _qj(k), such that t _rj(k) = t _pj(k)+ t _qj(k). As discussed previously, the autosampler procedure r_j(k) will generally comprise an additional time components inject, which is the time require to dispense the sample and t _zj(k), which is the time associated with post-injection operations zj(k).
• The previous discussion has noted that the maximum encountered value of the total pre-injection time may be stored in a database and used to calculate autosampler lag times and their corresponding delay times. It may only be necessary to store time values relating to the total pre-injection time, provided that autosampler lag times are measured from the time of completion of sample injection into the LC system. However, it is nonetheless also possible to maintain separate records of the component values t _pj(k) and t _qj(k) and, possibly, inject, or t _zj(k). The timing database may maintain records of the actual time intervals or, alternatively, of one or more quantities derived from the time values, such as a maximum value, minimum value, average value or standard deviation of values. The records relating to the pre-sample time component, t _pj(k) and the sample transport time component t _qj(k) may be useful if a system error occurs during processing of a sample, since the best error recovery procedure may depend on whether the autosampler was performing pre-sample operations or transport operations at the time of the error. In case of an error, the elapsed time since the most recent "Ready" System Sync signal may be compared to the time values in the database and used to determine the processing stage of the autosampler at the time of the malfunction. For instance, if a malfunction occurs in a detector component of an LC system while the autosampler is performing pre-sample operations, then the autosampler control process module may simply abort the currently executing steps and re-start operations when the detector is once-again functional since, in this case, the sample on which an analysis was to be performed remains undisturbed in its original container. However, if the same error occurs during the sample transport stage, then the sample may need to be discarded, since the integrity of the sample and the accuracy of analytical results obtained from it are generally at risk once a portion of sample has been withdrawn. In a multiplexing system that includes a single autosampler that can perform an injection into any one of multiple LC channels via a respective injection valve (e.g., see FIG. 3), then if a malfunction occurs in a channel that was to be the destination of a sample already in transport in the autosampler, the autosampler may be directed to wait for next availability of one of the still-functioning LC channels and to re-direct the sample injection to that channel.
• In accordance with the discussion above, FIG. 8A is a flow diagram of a first method, method 210, in accordance with the present teachings, for optimizing the operation of an autosampler and an LC system. The first step, Step 211, is an initialization step in which the Current Greatest Pre-injection Time for the n^th analytical procedure, in other words, the maximum previously encountered value of the time required for the autosampler pre-injection operations, $t_{\mathrm{rn}}^{\max },$

  

  is set to zero and stored in the timing database. The n^th analytical procedure is the next analytical procedure (in a sequence of analytical procedures) to be run by the system. The index "n" (where 1≤n≤N) is provided to identify a particular type of analytical procedure or procedural format - for instance, a type of sample analysis such as a particular type of assay - out of a total of N possible types of analytical procedures or procedural formats which may be executed by the system. It is possible that N=1 in which case the system always executes one type of analytical procedure.
• Subsequent steps 213-241 comprise a main iterated loop of the method 210 (FIG. 8A). In Step 213, timing data is input or read from the timing database. This data that is input will at least include the most recent value of $t_{\mathrm{rn}}^{\max }\mathrm{.}$

  

  Of course, a different value of $t_{\mathrm{rn}}^{\max }$

  

  will generally be stored in the database for each value of the index n (where 1≤n≤N). The input data may also possibly include the value of the required sample analysis time, t _sm, for the currently executing LC analytical procedure, generally the m^th analytical procedure and the starting time t _start of that procedure. The currently executing analytical procedure (the m^th procedure) may or may not be the same analytical procedure as the procedure (the n^th procedure) for which the autosampler is conducting preparatory steps. The quantity t _sm may be a constant, or alternatively, it may comprise a measured value obtained from previous execution of the m^th procedure by the system or may comprise a derived quantity such as a minimum observed value, maximum observed value, average value, etc. The quantities $t_{\mathrm{rn}}^{\max }$

  

  and t _sm are used, in the next step, Step 215, to set an autosampler lag time, Δt _rn, which is calculated as ${\mathrm{\Delta }\mathit{t}}_{\mathrm{rn}}^{\mathrm{sm}}=\left(t_{\mathrm{sm}}-t_{\mathrm{rn}}^{\max }\right)\mathrm{.}$

  

  In the next step, Step 217, the autosampler waits for a period of time corresponding to the calculated autosampler delay time, if the calculated autosampler delay time is greater than zero. The delay time, Δt _d, that corresponds to the lag time may be calculated from ${\mathrm{\Delta }\mathit{t}}_{\mathrm{d}}=\left(t_{\mathrm{start}}+{\mathrm{\Delta }\mathit{t}}_{\mathrm{rn}}^{\mathrm{sm}}\right)-t$

  

  in which tstart is the clock time (actual or system time) that was recorded (e.g., in Step 243) when the currently executing LC analysis procedure started and t is the current clock time.
• The autosampler commences its pre-injection sequence of operations in Step 225 of the method 210 (FIG. 8A). However, prior to doing so, the time at the commencement of the autosampler operations - for instance, a system time - is noted in Step 219 in order to keep track of the subsequent time required to perform the pre-injection operations. Alternatively, a timer (pre-injection timer or PI Timer) may be started in Step 219 such that the timer might operate as a stopwatch. In Step 231, after execution of the autosampler pre-injection steps, note is made of the actual time at the completion of the autosampler steps and a time difference is calculated. Alternatively, the dedicated timer, if utilized, is stopped. The program variable PI TIMER (or some other name) stores this time difference or Timer value. In Step 233, the value of PI TIMER is compared to the maximum previously encountered value of the time required for the autosampler pre-injection operations, $t_{\mathrm{rn}}^{\max }\mathrm{.}$

  

  If the value of PI TIMER is the greater quantity, then $t_{\mathrm{rn}}^{\max }$

  

  is reset to the value of PI TIMER, which new value is stored in the database, in Step 235.
• In Step 237 of the method 210 (FIG. 8A), a check is made to determine if the LC system is ready to receive the aspirated sample from the autosampler (e.g., if an LC ready System Sync signal has been transmitted or set). If not, the autosampler waits, in Step 239, for the receipt of such a signal. Once the LC system is ready, then, in Step 241, the sample is injected into the LC system and a signal (e.g., an Inject signal) may be transmitted or set so as to notify the LC system to begin the sample analysis operations.
• The steps outlined in solid-line boxes in FIG. 8A may generally be carried out by an autosampler control process (e.g., process 204 in FIG. 6) in a multiprocessing system that also includes an LC control process (e.g., process 208 in FIG. 6). Accordingly, after step 241 of the method 210 is executed, the in-process flow loops back to Step 213. However, several steps from a concurrently executing LC control process are shown in boxes with dotted lines in FIG. 8A, since these steps are triggered by the sample injection and transmission or setting of an Inject signal in Step 241. Thus, in Step 245 of the method 210, the LC control process executes all of the various steps involved in performing the analysis of the recently injected sample. Prior to doing so, however, note is optionally made (Step 243) of the actual or system time (or a timer is started) so as to keep track of the total time required, t _sn, to perform the n^th sample analysis procedure. This time is the total time between receipt of an Inject signal by the LC system and the transmission of a next LC Ready signal by the system. In Step 247, note is made of the actual or system time at the end of the LC analysis procedure and a time difference is calculated. Alternatively, the dedicated timer, if utilized, is stopped.
• It may be useful to maintain, within the timing database, a record of the sample analysis time for the m^th procedure, t _sm, corresponding to each execution of the analysis procedure. Alternatively, a record may be retained only for a derived quantity, such as, for instance, an average of several measurements of t _sm or a greatest encountered value or a minimum encountered value of t _sm. Of course, a different value of t _sm will generally be stored for each value of the index m (where 1≤m≤N). Such records may be useful because, as indicated in FIG. 7A, the autosampler delay time calculation takes this quantity into account. Although chromatography experiments are generally performed under tight constraints designed to maintain the experiments of constant duration, it is possible that the sample analysis time may vary over the course of several LC procedures. If this happens, a database record of such variation can enable continuous calculations of optimal delay times. The current value of t _sm or of a quantity derived therefrom is thus optionally stored in the database in Step 249 of the method 210. If it is desired to ensure that the autosampler module will almost always commence operations early enough so as to complete its pre-injection tasks prior to the time at which the LC module is ready to receive a sample, then it may be desirable, in this step, to compare the most recent value of t _sm to a minimum observed value of this quantity and to replace the previous minimum with the most recent value of t _sm if this value is less than the previous minimum. The LC Ready signal is set in Step 251.
• FIG. 8B is a flow diagram illustrating a portion of an alternative method, method 260, in accordance with the present teachings. Steps 211-217 and Steps 233-251 of the method 260 (FIG. 8B) are identical to the corresponding steps of the method 210 (FIG. 8A) and, therefore, these steps are not reproduced in FIG. 8B. The illustrated steps 218-229 of the method 260 replace the Steps 219, 225 and 231 previously shown in FIG. 8A. The method 260 differs from the method 210 through the provision (in method 260) of keeping separate records of the pre-sample time, t _pn, (Step 222) and the sample transport time, t _qn, (Step 227) relating to the n^th sample analysis procedure.
• The quantities t _pn(k) and t _qn(k) shown in FIG. 8B are the values of pre-sample time and sample transport time observed for the k^th execution of the n^th procedure. It is possible to store records (Step 228) comprising a plurality of measurements these quantities or else to store derived quantities, such as an average, a maximum observed value, a minimum observed value, etc. However, the Pre-Injection Timer is the sum (t _pn(k) + t _qn(k)) and is determined on every iteration of the steps shown in FIG. 8B (e.g., see Step 229 of the method 260), since this value is, in each iteration, compared to a prior maximum value.
• For use with a multiplexed LC system comprising several channels, such as the four channels illustrated in FIG. 3, the method shown in FIG. 8A may be slightly modified. A flow diagram of an appropriately modified method, in accordance with the present teachings, is illustrated in FIG. 8C. Steps 211, 217-237 and 245-251 of the method 280 illustrated in FIG. 8C are identical to the correspondingly numbered steps of the method 210 illustrated in FIG. 8A and are thus not re-described. However, the previously described Steps 213, 215, 239, 241 and 243 are modified, in FIG. 8C, to Steps 213b, 215b, 239b, 241b and 243b and a new Step 214 is inserted between Step 213b and Step 215b.
• The first step, Step 211, of the method 280 is an initialization step in which, as in the earlier described method 210, the Current Greatest Pre-injection Time for the n^th analytical procedure, $t_{\mathrm{rn}}^{\max },$

  

  is set to zero and stored in the timing database. However, in the next step, Step 213b, the method retrieves the remaining sample analysis time t _rem(i), for each ith chromatographic channel. For example, considering the system 50 illustrated in FIG. 3 as a multiplexed LC system, it may be observed that the system comprises four parallel channels, each channel including a pump 56, an injection valve 54 fluidically coupled to the pump and a column 58 fluidically coupled to the injection valve. Thus, when utilized in conjunction with the multiplexed system 50, step 213b would retrieve the four values t _rem(1), the time remaining to complete the sample analysis procedure currently being executed on the first channel, as well as t _rem(2), t _rem(3) and t _rem(4), which are the times similarly remaining with regard to the second, third and fourth channels. These time-remaining values may be determined from the formula t _rem(i) = t _s(i) - [t - t _start(i)] where t _s(i) is the estimated total required sample analysis time for the sample analysis procedure currently being executed on the i^th channel, t _start(i) is the time at which that analysis procedure was begun (stored in memory or a database in Step 243b or, optionally, in Step 241b) and t is the actual clock time or system time. Alternatively, an estimated analysis-end time, t _end(i), for each channel may be stored in memory or a database in Step 243b or, optionally, in Step 241b and the time-remaining values may be determined from the formula t _rem(i) = t _end(i) - t.
• In the next step, Step 214, of the method 280 (FIG. 8C), a particular channel is chosen to receive the injection of the next sample, the chosen channel, having channel index, c, being the one that will be available at the soonest time - that is, the channel associated with the minimum remaining time to completion of the currently executing sample analysis procedure. In the next step, Step 215b, the autosampler delay time is calculated using the time-remaining value for the soonest-available channel. Later in the method, in Step 241b, the sample aspirated by the autosampler is injected into channel c, possibly after waiting for this channel to be available in Step 239b.
• The discussion included in this application is intended to serve as a basic description. Although the present invention has been described in accordance with the various embodiments shown and described, one of ordinary skill in the art will readily recognize that there could be variations to the embodiments and those variations would be within the scope of the claimed invention. The reader should be aware that the specific discussion may not explicitly describe all embodiments possible; many alternatives are implicit.

Claims (10)

1. A system comprising an autosampler (2), a liquid chromatography (LC) apparatus (9), an electronic memory storage device (209) electronically coupled to both the autosampler (2) and the LC apparatus (9), a data line (210) electrically coupled between the autosampler (2) and the LC apparatus (9), an autosampler control process (204) and an independently operating LC control process (208), wherein the LC control process (208) is operable to cause the LC apparatus (9) to chromatographically separate components of a sample in accordance with one or more LC analysis procedures and to transmit a signal to the autosampler control process upon completion of each LC analysis procedure, wherein the autosampler control process is operable to cause the autosampler to delay operations for a delay time, Δt _d, after receipt of the signal, withdraw a portion of a first sample from a container, prepare the portion of the first sample for injection into the LC apparatus, transport the portion of the first sample to the LC apparatus and inject the portion of the first sample into the LC apparatus in accordance with the one or more LC analysis procedures, the system CHARACTERIZED IN THAT the autosampler control process (204) is operable to:

   read, from a database stored on the electronic memory storage device (209), a stored value of a pre-injection time, $t_{r1}^{\left(\max \right)},$

   

   that is required for the autosampler to perform operations preparatory to injecting the portion of the first sample into the LC apparatus (9) for analysis by a first LC analysis procedure, wherein the database includes a plurality of stored pre-injection time values, $t_{\mathrm{r}1}^{\max },$

   

   $t_{\mathrm{r}2}^{\max },$

   

   $t_{\mathrm{r}3}^{\max },$

   

   ... that correspond to respective different LC analysis procedures;

   calculate the autosampler delay time, Δt _d, from the stored pre-injection time value, $t_{\mathrm{r}1}^{\max },$

   

   and a time, t _s2, required to perform LC analysis of a second sample;

   cause the autosampler to perform the autosampler operations preparatory to injecting the portion of the first sample into the LC apparatus after delaying said autosampler operations for a time corresponding to the autosampler delay time, Δt _d; and

   if no autosampler or LC-apparatus errors or malfunctions have been encountered during the performing of said autosampler operations, cause the autosampler to inject the portion of the first sample from the autosampler into the LC apparatus after receipt of the signal from the LC apparatus.
2. A system as recited in claim 1, wherein the autosampler control process (204) is operable to determine the stored value of the pre-injection time, $t_{\mathrm{r}1}^{\max },$

   

   value by calculating a sum of a stored value of a pre-sample time, t _p1, that corresponds to autosampler operations associated with preparing to withdraw the portion of the first sample from a container and a stored value of a transport time, t _q1, that corresponds to autosampler operations associated with transporting and/or preparing of the portion of the first sample.
3. A system as recited in claim 1, wherein the autosampler control process (204) is further operable to:

   measure an elapsed time value, t _r1(i), corresponding to the step of performing the autosampler operations preparatory to injecting the portion of the first sample into the LC apparatus (9); and

   replace the value of the stored pre-injection time, $t_{\mathrm{r}1}^{\max },$

   

   stored in the database on the electronic memory storage device (209) with the measured elapsed time value, t _r1(i), if the measured elapsed time value is greater than the value of the stored pre-injection time, $t_{\mathrm{r}1}^{\max }\mathrm{.}$

   
4. A system as recited in claim 2, wherein the autosampler control process (204) is further operable to:

   measure a first elapsed time value, t _p1(i), corresponding to the autosampler operations associated with preparing to withdraw the portion of the first sample from the container;

   measure a second elapsed time value, t _q1(i), corresponding to the autosampler operations associated with transporting of the portion of the first sample or with preparing and transporting of the portion of the first sample;

   replace a stored value of pre-sample time stored in the database on the electronic memory storage device (209) with the measured first elapsed time value, t _p1(i), if the measured first elapsed time value, t _p1(i), is greater than the value of the stored pre-sample time; and

   replace a stored value of transport time stored in the database on the electronic memory storage device (209) with the measured second elapsed time value, t _q1(i), if the measured second elapsed time value, t _q1(i), is greater than the value of the stored transport time.
5. A system as recited in either of claims 2 or 4, wherein the autosampler control process (204) is further operable to abort execution of a currently executing autosampler operation and any subsequent autosampler operations associated with the portion of the first sample if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations associated with preparing to withdraw the portion of the first sample from the container.
6. A system as recited in either of claims 2 or 4, wherein the autosampler control process (204) is further operable to discard the portion of the first sample if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations associated with transporting of the portion of the first sample or with preparing and transporting of the portion of the first sample.
7. A system as recited in any one of claims 1-4,
   wherein the LC apparatus comprises a plurality of channels; and
   wherein the autosampler control process (204) is further operable to, if an error or malfunction of the LC apparatus is encountered during the performing of the autosampler operations preparatory to injecting the portion of the first sample into the LC apparatus, delay autosampler operations associated with transporting of the portion of the first sample or with preparing and transporting of the portion of the first sample until one of the plurality of channels is available to receive the sample portion.
8. A system as recited in claim 1, wherein the autosampler control process (204) is further operable to:

   read, from the database on the electronic memory storage device (209), the time, t _s2, required to perform LC analysis of the second sample prior to the calculating of the autosampler delay time;

   measure an elapsed time value required to perform LC analysis of the second sample; and

   replace the stored value, t _s2, stored on the electronic memory storage device (209) with the measured elapsed time value required to perform LC analysis of the second sample if the measured elapsed time value required to perform LC analysis of the second sample is less than the stored value, t _s2.
9. A system as recited in any one of claims 2, 4, 5 or 6 wherein the stored pre-sample time value, t _p1, comprises one of a plurality of stored pre-sample time values, t _p1, t _p2, t _p3, ..., each of the plurality of stored pre-sample time values corresponding to autosampler operations associated with preparing to withdraw a sample portion from a container in accordance with a different respective one of a plurality of LC analysis procedures.
10. A system as recited in any one of claims 2, 4, 5, or 6 wherein the stored transport time value, t _q1, comprises one of a plurality of stored transport time values, t _q1, t _q2, t _q3, ..., each of the plurality of stored transport time values corresponding to autosampler operations associated with transporting of a sample portion or with preparing and transporting of the sample portion, in accordance with a different respective one of a plurality of LC analysis procedures.

Images